JPH0822405A - Management system for relational data base - Google Patents

Abstract

PURPOSE:To understand the relational data base as three-dimensional structure and to enable generation management by record by incorporating the concept of records in the relational data base. CONSTITUTION:To alter route data, an alteration instruction for the route data is inputted on the keyboard 12 of a personal computer 1 together with a valid date on the basis of the alteration specifications of the route data. Consequently, a control part 11 retrieves a reference consistency table 21 to confirm whether or not an object attribute has parent-child relation. As a result of this confirmation, it is judged that the route name of the route data in a relational data base 22b as a parent attribute and the route name of station data in a relational data base 22a as a child attribute are altered. According to this judgement, the control part 11 updates the tuple of the corresponding route data in the relational data base 22b and the tuple of the corresponding station data in the relational data base 22a on the basis of the valid date.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a relational database management system for managing a relational database having referential integrity and performing generation management.

[0002]

2. Description of the Related Art Conventionally, equipment such as commuter ticket issuing machines and automatic ticket gates have been used in transportation systems such as railways. These devices manage route data and the like based on data files stored in the internal hard disk.

However, the route name of a certain company line is changed, a new station is added, or a station is abolished. In this case, not only the data file of the device for the company line, but also the data file of each device of the other company's line that is in contact is changed.

A relational database in which relations such as lines, companies, stations, and crossing stations are prepared is used as a database on which this change is made. As a data updating method that does not impair referential integrity in this relational database, the relational database language SQL is standardized by the ISO standard (JIS standard). However, in a generation management database that performs generation management on a record-by-record basis, generation management is realized by handling multiple tuples with valid periods set as one record, so to maintain referential integrity The referential constraint operation is not performed individually for all generations that overlap the valid period of the symmetric tuple.

For this reason, SQL cannot be directly applied,
The database user had to check the referential integrity for one data update and execute multiple instructions. Therefore, in the conventional generation management database,
Since referential integrity is not guaranteed, data reliability is poor, and it is difficult to use a database without knowledge of generation management and referential integrity. Further, since a plurality of instructions are required to update the data, the amount of work is increased, and the application program using the database is inevitably complicated.

[0006]

As described above, the present invention eliminates the drawback that referential integrity cannot be realized in a relational database that manages generations in record units. In a relational database that manages generations,
It is an object of the present invention to provide a relational database management system that can realize referential integrity.

[0007]

The relational database management system of the present invention is a reference storing the referential integrity of a plurality of three-dimensional relational databases consisting of attributes, tuples, and effective dates. A consistency table, a designating means for designating the update content of the relational database and its effective date, and the stored content of the target relational database based on the update content designated by the designating means based on the valid date. A process for updating, determining a referential integrity relational database from the referential integrity table based on the updated content, and updating the stored content of the determined referential integrity relational database based on the valid date. Is composed of means

A relational database management system according to the present invention is a three-dimensional relational database consisting of attributes, route names, and effective dates of the route names, and a referential integrity storing the referential integrity of each of the relational databases. Property table, specifying means for specifying the update contents of the relational database and the effective date thereof, and updating the stored contents of the target relational database based on the update contents specified by the specifying means on the basis of the effective date. Processing means for determining a relational database having referential integrity from the referential integrity table based on the update content and updating the stored content of the determined relational database having referential integrity based on the valid date. It consists of

[0009]

According to the present invention, by introducing the concept of a record into the relational database in the above structure, the relational database is understood as a three-dimensional structure, and generation management for each record is realized. is there.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows a management system for managing data files of transportation equipment in which the relational database of the present invention is used.

In this management system, the stored contents of each data file are changed by a personal computer 1 for a database administrator and a floppy disk 2 in which a data file created by the personal computer 1 is stored. It is composed of station affairs equipment 3, ...

As shown in FIG. 2, the personal computer 1 includes a control unit 11 for controlling the whole, a hard disk HD in which various relational databases are stored,
A keyboard 12 for inputting various instructions, a display 13 for guidance, a data conversion unit 14 for converting the stored contents of each relational database into data corresponding to the contents of each station service device 3, ... Converted data to each floppy disk 2, ...
The floppy disk device 15 stores the data in

In the hard disk HD, the referential integrity table 21 and the relational database 2 are stored.
2a, 22b, ... Are provided. The referential integrity table 21 includes the relational database 22a, ...
It is possible to define referential integrity of and to associate one or more attributes called foreign keys in one relational database with the attributes of another relational database.

The referential integrity table 21 is a table in which the relationship between the parent attribute and the child attribute as the correspondence is stored. For example, as shown in FIG. 3, the relation name and attribute name of the child attribute and the parent attribute are stored. Relation names, attribute names, and their relationships are stored correspondingly.

This relationship conforms to the referential integrity standardized by SQL standardized by ISO, and by defining the relevance to the attributes of two relations, the organic relations of a plurality of relations are defined. It guarantees a strong connection.

For example, in the case of FIG. 3, the child attribute relation name "station" and the attribute name "route name", the parent attribute relation name "route" and the attribute name "route name", and the relationship "match" correspond to each other. Remembered

It is impossible to update a record that cannot maintain the referential integrity, and it is an error or is it possible to propagate the update to other relational databases so as to maintain the referential integrity? The database user can select it.

The relational database 22a,
Each is a generation management type relational database, and can be regarded as three-dimensional data incorporating the effective period (see (a), (b), and (c) in FIG. 4).

These relational databases 22
In a, ..., by introducing the concept of a record as a set of tuples corresponding to several consecutive effective periods, it is treated as if the conventional two-dimensional relational data changes according to a specified date. You can

That is, FIG. 4 (a) shows the two-dimensional data of a conventional relational database to which height as a time axis is added. The same two-dimensional data as the relational data in which the horizontal axis has attributes and the vertical axis has tuples are obtained (see (c) of FIG. 4). At this time, some attributes specified as the primary key do not overlap with other tuples.

On the other hand, when the data is cut out in the vertical direction, two-dimensional data showing the change of the corresponding record due to the date is obtained (see (b) in FIG. 4). This time,
All primary keys are the same. When an update command is executed for a record, this table is rewritten at a designated date so that the update content is reflected in a two-dimensional table cut out in the horizontal direction.

The relational database 22a is for station data, for example, and as shown in FIG. 4A, each axis is composed of attribute information, tuple numbers, and effective dates. In this case, the attribute information is the start date,
It consists of abolition date, station code, route name, station name, and display pattern.

For example, for a station code "4" as a tuple number, a tuple group (record) as shown in FIG. 4B is stored.
And tuples with different display patterns (there are no tuples with overlapping dates).

In this case, the start date "90-01-01",
Abolition date "91-12-31", station code "4", line name "Chuo Line", station name "Shinjuku", and tuple of display pattern "Shinjuku", start date "92-01-01", abolition date " 93-02-01 ", station code" 4 ", route name" Chuo Line ", station name" Shinjuku ", and tuple of display pattern" Shinjuku ", start date" 93-02-02 ", abolition date" 94- "
2-16 ”, station code“ 4 ”, route name“ Chuo Line ”, station name“ Shinjuku ”, and tuple of display pattern“ Shinjuku ”,
The tuple includes a start date “94-02-17”, an abolition date “NULL”, a station code “4”, a route name “Chuo Line”, a station name “Shinjuku”, and a display pattern “XX Shinjuku”.

Further, a tuple group as shown in FIG. 4C, which is effective for 94-02-17, corresponds (the same station code (key) does not exist). In this case, start date "94-01-01", abolition date "NULL", station code "1", route name "Tokaido Main Line", station name "Tokyo", and tuple of display pattern "○○ Tokyo", start date "93-
01-31 ", abolition date" 94-02-28 ", station code" 2 ", line name" Tokaido Main Line ", station name" Yokohama ", and tuple of display pattern" Yokohama ", start date" 93-0 "
9-01 ”, abolition date“ NULL ”, station code“ 3 ”, line name“ Tohoku Main Line ”, station name“ Sendai ”, and display pattern“ Sendai ”tuple, start date“ 94-02-17 ”,
It is composed of tuples of abolition date "NULL", station code "4", line name "Chuo Line", station name "Shinjuku", and display pattern "XX Shinjuku".

Further, the relational database 22b
Is for route data, for example, and each axis is composed of attribute information, tuple numbers, and effective dates. In this case, the attribute information includes a start date, an abolition date, a route code, a route name, a company name, and a display pattern.

That is, tuple groups as shown in FIG. 5, which are effective for 94-04-01 and later, correspond (the same route code (key) does not exist). In this case, start date "93-01-01", abolition date "NULL", route code "1", route name "Tokaido Main Line", company name "○○○"
○○ ”and tuple of display pattern“ Tokaido Main Line ”,
Start date "94-04-01", abolition date "NULL", station code "2", route name "Tohoku Main Line", company name "○○○○"
○, and tuple of display pattern “Tohoku Main Line”, start date “90-01-01”, abolition date “NUL
L ”, route code“ 3 ”, route name“ Chuo Line ”, company name“ ○○○○○ ”, and display pattern“ Chuo Line ”
It consists of tuples.

Therefore, the control section 11 creates a data file of the designated date for each station service device 3, ... In accordance with the stored contents of the relational databases 22a, 22b ,. It is stored in the floppy disk 2, .... By setting each of the floppy disks 2, ... Into the corresponding station service device 3 ,.
The data file of the hard disk (not shown) in ... is updated with new contents.

Next, the process of changing the route data of the company A in the above-mentioned configuration will be described with reference to the flowchart shown in FIG. First, a route data change command is input using the keyboard 12 of the administrator's personal computer 1 based on the A company route data change specification. This route data change command is, for example, the relation name “route data”, the key name “3”, the change attribute “route name”, the change content “Chuo Line” is “Chuo Main Line”, and the change date is “94-01-01”. It has become.

As a result, the control unit 11 searches the referential integrity table 21 and confirms whether or not the target attribute has a parent-child relationship. As a result of this confirmation, it is determined that the relation name "route" and the attribute name "route name" are parent attributes, and the relation name "station" and the attribute name "route name" are child attributes to this parent attribute. That is, it is determined whether the line name “Chuo Line” in the line data of the relational database 22b and the line name “Chuo Line” in the station data of the relational database 22a are changed.

Based on this judgment, the control unit 11 updates the tuple of the corresponding route data in the relational database 22b and the tuple of the corresponding station data in the relational database 22a.

That is, with respect to the route data shown in FIG. 7A, the abolition date of the third tuple is changed to "NULL" as shown in FIG. , Start date "90-01-01", abolition date "NUL
L ”, route code“ 3 ”, route name“ Chuo Line ”, company name“ ○○○○○ ”, and display pattern“ Chuo Main Line ”
Add a tuple of.

Further, with respect to the station data shown in FIG. 8A, the abolition date of the third tuple is changed to "93-12-31" as shown in FIG. The route name of the second tuple was changed to "Chuo Main Line", and the start date "94-01-01" and the abolition date "94" were set between the third and fourth tuples.
-02-16 ”, station code“ 4 ”, line name“ Chuo Main Line ”, station name“ Shinjuku ”, and display pattern“ Shinjuku ”tuples are added.

As a result, as shown in FIGS. 9A and 9B, each tuple of the route data and the station data is 94-01-
Based on 01, the route name can be changed to "Chuo Main Line", and their consistency can be maintained.

In the above example, the change has been described, but the abolition, addition, and the like can be similarly performed. For example, if the route name “Chuo Line” is to be abolished by adding 94-01-01, the route data shown in FIG.
The abolition date of the third tuple was changed to "94-01-01", and the abolition date of the third tuple was changed to "94-01-01" for the station data shown in Fig. 8A. Delete the fourth tuple. Also, the addition is to add a tuple as a specified description, and only confirms whether or not it can be done by date. Records that violate referential integrity cannot be created during this addition.

Therefore, in the above generation management type relational database, when a record is updated, not only a command that destroys referential integrity is rejected but also another relation is maintained to maintain referential integrity. The user can specify that the changes in This time,
Each record can be updated so that referential integrity is maintained at any date, not limited to the change of tuples standardized in SQL2.

Further, since the relational database manages generations, it is possible to change data that must be performed on a specific date in advance without impairing referential integrity.

Further, by applying the referential integrity to the relational database of the generation management type, the referential constraint operation can be designated, the reliability of the data is improved, and the amount of data updating work is reduced.

Furthermore, by allowing the relational database to manage the start date, the abolition date, and the management of referential integrity, tuples of each generation having the same content are regarded as records, and an update command is issued to the records to generate a plurality of tuples. Make changes easily and reliably.

[0040]

As described above in detail, according to the present invention,
A relational database management system capable of realizing referential integrity in a relational database that manages generations in record units can be provided.

[Brief description of drawings]

FIG. 1 is a schematic diagram for explaining the configuration of a relational database management system according to an embodiment of the present invention.

FIG. 2 is a block diagram showing an internal configuration of a personal computer.

FIG. 3 is a diagram showing a storage example of a referential integrity table.

FIG. 4 is a diagram for explaining a relational database for station data.

FIG. 5 is a diagram for explaining tuple groups for route data.

FIG. 6 is a flowchart illustrating a route data change process.

FIG. 7 is a diagram for explaining changes in route data.

FIG. 8 is a diagram for explaining changes in station data.

FIG. 9 is a diagram for explaining the consistency of each tuple of route data and station data.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Personal computer 2 ... Floppy disk 3, ... Station work equipment 11 ... Control part HD ... Hard disk 22 ... Reference consistency table 22a, 22b ...

Claims (1)

[Claims] 1. A plurality of three-dimensional relational databases consisting of attributes, tuples, and effective dates, a referential integrity table storing referential integrity of each of the relational databases, an update content of the relational database, and its contents. Based on the effective date, the storage means of the target relational database is updated based on the update means specified by the specifying means and the update content specified by the specifying means, and the reference matching is performed based on the update content. A relational database having a referential integrity is determined from the compatibility table, and processing means for updating the stored content of the determined referential integrity relational database based on the valid date is provided. Database management system. 1. A plurality of three-dimensional relational databases composed of attributes, route names, and effective dates of the route names, a referential integrity table storing referential integrity of each of the relational databases, and the relational database. Specifying means for designating the update contents and the effective date thereof, and based on the update contents designated by the specifying means, the stored contents of the relational database to be updated are updated based on the effective date, and Based on the referential integrity table, a relational database having referential integrity is determined based on the validity date, and the stored content of the determined relational database having referential integrity is updated based on the valid date. A relational database management system.